Crossing the street is a complex, difficult and dangerous task. When making street-crossing decisions, pedestrians must allow for a preferred safety margin and judge whether the gap in vehicular traffic exceeds the time needed to cross the street. Previous street-crossing research has shown that the elderly, visually impaired, and blind pedestrians make unsafe street-crossing decisions. This study will evaluate the street-crossing decision making performance of these high-risk pedestrian groups. We will use an innovative criterion-free metric to evaluate the accuracy and reliability of pedestrians' judgments on whether the vehicular gap time is of sufficient duration to allow them to cross the street. The pedestrian's preferred safety margin will not be included in this crossing decision. We will also measure each pedestrian's criterion for crossing by determining the gap time when each pedestrian first thinks that it is safe to cross. The pedestrian's safety margin in this crossing decision will be included. The collected data will address four aims. Aim 1 will assess how street-crossing decision making performance changes as a function of vision (visual acuity, contrast sensitivity and visual fields), and a pedestrian's characteristics (age and walking speed). If pedestrians do not compensate for their sensory and motor limitations when making their street-crossing decisions, we expect to see a corresponding decrease in their accuracy and/or reliability and a shortened criterion for crossing. If pedestrians compensate for their limitations and adopt a conservative or cautious street-crossing strategy the converse will be found. Aim 2 will compare the accuracy, reliability and criterion for crossing between visually impaired and blind pedestrians who self-report having difficulty crossing the street and those who self-report having no difficulty. If street-crossing decision making performance is related to a pedestrian's self-confidence, those pedestrians who self-report having difficulty would exhibit unsafe street-crossing decisions. Pedestrians who have no difficulty would exhibit safe street-crossing decisions. Aim 3 will test the hypothesis whether pedestrian accuracy, reliability and the criterion for crossing improves after training with a current Orientation and Mobility (O&M) street-crossing decision training program. Aim 4 will compare the accuracy, reliability and criterion for crossing in pedestrians measured at three different types of crossing environments. Consistent street-crossing decision making behavior across different environments would suggest that the study's findings are generalizable. Inconsistent decision behavior would be informative and warrant further investigation. Collectively, the project's findings will provide a more detailed understanding of the factors that affect street-crossing decision making performance. They will also guide the development of objective evidence-based recommendations for low vision (LV) rehabilitation specialists about when they should refer patients for street-crossing decision training.